113 - Don't overthink AI, get started
Where should you start with using AI - it doesn't matter just start. Marriott becomes an ad-platform. The 95-5 rule in b2b marketing and more.
Hello,
Heading to HITEC for the last day in a few minutes. There’s a lot of AI-talk. Sometimes (unfortunately) it is fluff. If you’re a vendor just remember this, people wont buy your products because it has AI, they buy your products because it solves their problems and if it had AI or not, matters less. (I know, investors want to see AI everywhere but customers don’t).
Best, Martin
Sponsored by Klairhaus. Support the newsletter, treat someone to great office gadgets.
Everyone is becoming an ad platform
MARRIOTT MEDIA, is leveraging first-party data from its 237 million members Bonvoy program. To sell ads. A few years ago it was becoming increasingly obvious that every platform that had first party data was going to become an ad platform. It is the highest margin product of Amazon. Google has shown that it prints money. But when does it become too much? When companies start to make products that are worse because they need to support ads. The first page (if not two) of Amazon and Google are ads now…
MARRIOTT MEDIA NETWORK
Should B2B Advertising Focus More on Brand Building?
The 95-5 rule, suggests that most B2B buyers are not in the market at any given time. That only 5% of the market are actually buying and 95% are not. So effective advertising should focus on creating lasting brand impressions. Especially in hotel technology where cycles are long and nobody likes to replace systems. So while we want leads now - never forget that if you don’t work on the 95% - there will be no leads next year.
THE 95-5 RULE
About me: I'm a fractional CMO for large travel technology companies helping turn them into industry leaders. I'm also the co-founder of 10minutes.news a hotel news media that is unsensational, factual and keeps hoteliers updated on the industry. Voice AI for the front desk
There are several great chatbots both text and voice in the industry. But as I was saying last week, AI is great when everything is running smoothly. They answer the dozen or so standard questions and keep the phones free. My recommendation here is to ensure the handover to human is smooth because when things do not go according to plan, speaking to a machine is the worst.
VOICE AI FOR HOTELS
Enterprise technology
The hotel industry is often seen as the laggard when it comes to technology adoption (compared to airlines, cruise lines etc). But it seems there are even bigger issues in other parts of the travel industry. Some air traffic control systems still run with floppy disks (well it definitely makes it harder to hack). The US has the busiest and most impressive air transport system in the world. It is so surprising then that airports and air traffic aren’t more optimized and modern.
AIR TRAFFIC CONTROL MODERNIZATION
Analysis of marketing tech in travel
Not groundbreaking news, but this report highlights some of the shifts in the travel marketing world and what is to come. Interesting read, key points: Proactively engage travelers before they even ask, Unify fragmented platforms into seamless experiences, Dynamically price services with real-time AI.
MARTECH TRAVEL LOYALTY
Self-driving cars and travel
Something happens when transport enter an area. Economy grows, travel begins, people set up shops etc. Self-driving cars are slowly starting to expand their footprint. This will change the dynamics of large cities, public transport and how tourism. Every time it become easier to explore, more people want to explore and tourism grows. Could this help distribute it more evenly?
WAYMO TOURISM IMPACT
Podcast: I was invited on the Hospitality Daily Podcast and spoke about technology in hospitality, some thoughts on what wont change in hospitality, and why I co-founded 10minutes.news. Opinion
Thoughts from HITEC and AI? Just start
I’m quite sure that one day someone asked “So… where do we start with this computer thing?” The answer was: anywhere. Some people opened Lotus 1-2-3 some people did graphic design, some wrote a letter. The point wasn’t the starting point, it was the fact we’d started. Generative AI is the same seismic moment, only faster and louder, and the “where do I start?” needs to change to “what’s next”.
Pick any task and throw AI at it today. Turn your restaurant’s PDF menu into a clean csv upload for your POS. Map room-type codes between two hotel systems in five minutes. Feed the hotel guest emails and have it flag reservations, complaints, and the one VIP request that somehow arrived at 02:14 on a Sunday.
Yes, hallucinations exist. So did “fatal errors” on DOS. So, double-check sensitive outputs, build guardrails, move on. Waiting for the perfect guide, framework, or consultant contract only guarantees someone else learns faster while you’re still negotiating the SOW.
The magic happens when daily use compounds. Day one you ask for the Eiffel Tower’s height. Day seven you paste 3 000 words of messy meeting notes and get a bullet-proof action list. Day thirty you’re speaking instead of typing, building custom GPTs, and wondering why people still type. What looks like wizardry from the outside is really just the muscle memory of just trying doing and trying every day.
Force yourself to do something with AI every day. Proofread your email, write a journal, suggest an itinerary. Just force yourself to do something and get familiar with what is possible.
So close the tab, open ChatGPT, Gemini, Copilot, Claude and ask it to solve something trivial but annoying. That first imperfect result is the real on-ramp. Everything after that is just exploration.
This is the only way to go. We want to think up a perfect plan, it is reassuring. But just like with computers a long time ago - there is no perfect plan. Your use case is different from mine.
Great content, Martin!
Love it!